(keywords: Mahayana-sutralamkara, Mahayanasutralamkara)
A new English translation of the Mahāyāna-sūtrālaṃkāra came out last month (November, 2014): Ornament of the Great Vehicle Sūtras: Maitreya’s Mahāyānasūtrālaṃkāra with Commentaries by Khenpo Shenga and Ju Mipham, translated by Dharmachakra Translation Committee (Boston & London: Snow Lion, 2014). It was preceded by two other English translations of this text: (1) Mahāyānasūtrālaṃkāra by ‘Asaṅga,’ Sanskrit Text and Translated into English by Dr. (Mrs.) Surekha Vijay Limaye (Delhi: Sri Satguru Publications, 1992); and (2) The Universal Vehicle Discourse Literature (Mahāyānasūtrālaṃkāra), By Maitreyanātha/Āryāsaṅga, Together with its Commentary (Bhāṣya), By Vasubandhu, Translated from the Sanskrit, Tibetan, and Chinese by L. Jamspal, R. Clark, J. Wilson, L. Zwilling, M. Sweet, R. Thurman (New York: American Institute of Buddhist Studies, 2004). The new translation has been hailed as the most readable one now available. While readability is important, even more important is accuracy. It will be worthwhile to compare the existing translations using this criterion.
The 2014 English translation of the Mahāyāna-sūtrālaṃkāra is a translation of a translation, being made from the Tibetan translation in the Derge edition (Translators’ Introduction, note 12, p. 964), without reference to the Sanskrit original. This allows us to see how this text was understood in Tibet, as do the two accompanying commentaries written in Tibet in comparatively recent times. The value of this is that the Buddhist tradition has been lost in India, its homeland, for about a thousand years now. Thus, as I have noted elsewhere,1 the 1992 English translation made in India from the published Sanskrit text (without reference to the Tibetan translation) is quite unreliable. The obviously sincere and well-meaning translator acknowledges the help of her teacher and of her supervisor (Introduction, p. xxiii), who clearly were unfamiliar with the Buddhist teachings. The common Buddhist phrase, śaraṇa-gamana, “going for refuge,” is here translated as “recourse to surrender” (p. 24); the term pudgala, used throughout Buddhism to mean “person,” is here translated as used throughout Jainism to mean “matter” (e.g., pp. 244, 441, 447, etc.); the phrase giving the fundamental Buddhist doctrine, ātma-dṛṣṭi, “(false) view of self,” is here translated as “one’s own view point” (p. 69). The Tibetan tradition regards itself as having preserved the Indian tradition intact, giving the original meaning of the text unchanged. A welcome window into the Tibetan exegesis of this text is provided by the new translation. For the Mahāyāna-sūtrālaṃkāra itself, however, modern scholarship must question whether a translation of a translation, however competently done, can take the place of a translation of the original, competently done.
The prior 2004 English translation of the Mahāyāna-sūtrālaṃkāra was also made from the Tibetan translation, but was then clarified and corrected by comparison with the published Sanskrit text, and with some reference to the early Chinese translation. When a text goes from a language having a very large vocabulary, such as Sanskrit, into a language having a much smaller vocabulary, such as Tibetan, something is inevitably lost. While the Tibetan tradition has no doubt correctly preserved the meaning of the Sanskrit text in general, to expect it to have captured every particular is unrealistic. Therefore, the translators of the 2004 translation felt the need to utilize the Sanskrit text. Because the Tibetan vocabulary is smaller than the Sanskrit vocabulary, one Tibetan word must translate more than one Sanskrit word. For example, in verse 6.3d (Sanskrit edition and 2004 English translation) or 7.3d (Tibetan translation and 2014 English translation), the Sanskrit word dharmamayaḥ was translated by the Tibetan words chos kyi rang bzhin. The Sanskrit word dharma is always translated by the Tibetan word chos, and is here used in its meaning, “the elements of existence” or “phenomena.” The Tibetan word rang bzhin most often translates the Sanskrit word svabhāva, “inherent nature.” This allowed the Tibetan words to be understood as the very common phrase used in philosophy, “the inherent nature of phenomena.” Thus, the 2014 translation has: “This is the nature of phenomena.” Here, however, the Tibetan word rang bzhin translates the Sanskrit suffix, -maya, “consisting of.” The verse is talking about people (Skt. janaḥ, Tib. skye bo), saying that they “consist of phenomena”; it is not making a statement about the nature of phenomena. Accordingly, the 2004 translation has: “they [beings] . . . are objective,” where by “objective” we are to understand that they are “objects,” “things,” “phenomena” (dharma-s).
Similarly, when putting teachings into metrical verses, which the Mahāyāna-sūtrālaṃkāra is composed of, words must often be altered or substituted to fit the meter. In Sanskrit texts, these words are usually restored in the accompanying commentary. In the Tibetan translations of these metrical verses, syllables must often be dropped to fit the Tibetan meter, which is regulated by the number of syllables per line of verse. Sometimes these omitted syllables are ones that provide necessary information, such as the declension or number of a word. Declensional endings, separate syllables in Tibetan, tell the reader how to take the word in the sentence. Without them, the reader is left to guess at the construal and intended meaning. For example, in verse 6.6c or 7.6c, the Sanskrit word dharmeṣu (locative declension, plural number) was translated by the Tibetan word chos la (accusative, dative, or locative declension, singular number). The syllable showing the plural number (rnams) was dropped to fit the meter. This allowed the word dharma or chos to be taken in the Tibetan translation in the singular, as “the Dharma,” i.e., the Buddhist teachings, rather than as the dharma-s, the “elements of existence” or “phenomena” or “things.” Thus, the 2014 translation has: “The bodhisattva contemplates the Dharma in a most decisive way”; while the 2004 translation has: “a bodhisattva becomes decisive in her judgment about things.”
The Tibetan translations are deservedly renowned for their high degree of accuracy in following the Sanskrit originals very closely. The Tibetan translations are much more literal than the great majority of English translations today. This literal accuracy has resulted in the most precise transferal of a body of religious knowledge from one language to another known to history. Because the Tibetan translations follow the Sanskrit originals so closely, their style is closer to Sanskrit than to native Tibetan. This at times can present a challenge in understanding them, and in translating these translations into English. When the Sanskrit text is available, ambiguities in the Tibetan translation can usually be clarified by reference to it. For example, the 2014 translation of the Mahāyāna-sūtrālaṃkāra, made only from the Tibetan, erroneously has (verse 2.7): “Because of its [the Great Vehicle’s] vastness and profundity, maturation and nonconceptuality, its teaching is twofold.” The 2004 translation, clarified by comparison with the Sanskrit, has (verse 1.7, or verse 1.13 in the Sanskrit edition): “From the magnificent and the profound come evolutionary development and nonconceptual (wisdom). (The universal vehicle) teaches both, . . .” The verse does not say, “Because of its . . . maturation and nonconceptuality,” but rather speaks of its twofold teaching of vastness and profundity, saying that maturation comes from vastness, and nonconceptuality comes from profundity. This is unmistakable in the Sanskrit. Many more errors of this type could be cited, that would have easily been avoided by reference to the Sanskrit.
The long lost Sanskrit text of the Mahāyāna-sūtrālaṃkāra and its accompanying commentary (bhāṣya) was discovered in 1898 in Nepal by Sylvain Lévi. It was then edited by him and published in Paris in 1907 (posted on this website under “Sanskrit Texts”), followed by his pioneering French translation in 1911. Lévi’s edition was based on a transcript made for him of a single manuscript,2 a paper manuscript written in 1677-1678 as we now know,3 and such manuscripts are notoriously full of scribal errors. Lévi’s edition became the basis of the 1970 edition by S. Bagchi, helpful because it corrects many misprints and other errors in Lévi’s edition (see Bagchi’s forty-page corrigenda), and these two became the basis of the 1985 edition by Dwarika Das Shastri. Lévi’s edition also became the basis of the 1992 translation by way of Bagchi’s edition, and was the Sanskrit text used for comparison for the 2004 translation. Lévi made many corrections to his 1907 Sanskrit edition in his 1911 French translation, and in 1958 Gadjin Nagao published eleven pages of corrections to Lévi’s edition, including those made by Lévi.4 Nagao’s corrections were based primarily on the Tibetan and Chinese translations and on Sthiramati’s sub-commentary (in Tibetan translation), and also on two additional Sanskrit manuscripts that were brought to Japan and are kept in the Ryukoku University Library.5 Accordingly, the 2004 translation says that “There are three known Sanskrit texts of the MSA” (Introduction, p. xxxiii), and the 2014 translation repeats this, referring to “the three extant Sanskrit manuscripts” (Translators’ Introduction, note 12, p. 964). In fact, additional Sanskrit manuscripts of this text exist in the Nepal National Archives.6 In 1985, Naoya Funahashi published chapters 1, 2, 3, 9, and 10 of a much needed revised edition, based on these additional manuscripts, and in 2000, a revised edition of chapter 11.7
The 2004 translation is the result of a longstanding effort involving several scholars, who produced a completed draft already by the end of the 1970s. So by 1980 the Sanskrit text had already been compared. Thus, the corrections by Lévi (1911) and Nagao (1958, as well as his later personal input) were utilized, but the revised editions by Funahashi (1985, 2000) were not utilized. Nor were the many corrections that Lévi had written in his personal copy of his edition, published only in 2001 thanks to the efforts of Kamaleswar Bhattacharya, filling eight printed pages.8 In the last few years, Kazuo Kano has been publishing the edited Sanskrit text of eight folios of a very old palm-leaf manuscript of the Mahāyāna-sūtrālaṃkāra and bhāṣya found at the Ngor monastery in Tibet.9 While these include valuable corrections, they also show that the Sanskrit text we have, disregarding scribal errors, is essentially the same as the one translated into Tibetan long ago. The 2004 translators shied away from referring to the “Sanskrit original” (Preface, p. x), because of the many errors in the comparatively late Sanskrit manuscripts found in Nepal (on one of which Lévi’s edition was based), but we can now certainly do so. The Tibetan translation, too, has numerous scribal errors, as may be seen by comparison between the various Tengyur editions.
A translation of the very helpful Sanskrit commentary that accompanies the verses, the bhāṣya by Vasubandhu, is included in full in the 2004 translation (and also in the 1992 translation, but this translation is simply too unreliable to take into account). The Tibetan translation of this commentary was, in effect, abridged by Khenpo Shenga (1871-1927), and was thus partially included in the 2014 translation by way of his commentary. Thus, good explanations of the often too brief verses are found in both the 2004 and 2014 translations. Sometimes the verses are not explained (or not fully explained) in the accompanying commentary, which is comparatively brief, so a larger commentary must then be consulted. In India, this larger commentary is the sub-commentary by Sthiramati, so far still lost in Sanskrit, but preserved in its Tibetan translation. In Tibet, the larger commentary by Ju Mipham (1846-1912) drew heavily upon the commentary by Sthiramati. A translation of Mipham’s lengthy commentary is included in full in the 2014 translation, bringing the page count of this translation to 929 pages. For the 2004 translation, Lobsang Jamspal read through the entire Sthiramati sub-commentary and adapted that translation accordingly.
The dust jacket of the 2014 translation quotes scholars describing it as an “outstanding translation,” and saying that “the translators have rendered this text . . . into the most accessible and readable English now available.” This is a polite way of adverting to the English of the 2004 translation as being less accessible and readable. An American longtime Buddhist put it more bluntly in an email reply to me shortly after the 2004 translation was published: “You are too kind to Thurman. I am disgusted that he took the serviceable version by ?? (forgot which Tibetan did it) [Lobsang Jamspal] and plugged in his ‘evolution,’ ‘genius’ and other ludicrous thurmanisms. I have tried to read it, but simply do not know what many of the thurmanisms correspond to. So it sits on the shelf. Thirty years wait and this is what we get! And he has no Tibetan-Thurman glossary so one could match up his goofy translation choices.” The English terminology in the 2004 translation is avowedly experimental (Preface, p. x), and Thurman’s translation choices for these terms were mostly adopted later in the joint translation process. Besides “evolution” or “evolutionary action” for karma, “evolutionary maturity” for paripāka (translated as “full maturation” in the 2014 translation), and “genius” for dhīmat (a common epithet of a bodhisattva, translated as “wise individual” in the 2014 translation), the 2004 translation employs translations such as “addictions” (or “mental addictions”) for kleśa. This basic term in Buddhism had long been translated as “defilements,” and more recently as “afflictions” (or “mental afflictions”), as it is in the 2014 translation. The 2004 translation also switches back and forth between “his” and “her” pronouns throughout, even though the original text does not, in deference to modern sensibilities about respect to women. Even the title was a last-minute change, translating alaṃkāra as “literature” rather than as “ornament” (Introduction, p. xiii, fn. 3).
From Thurman’s lifelong work and publications, I have no doubt that his translation choices are motivated by the bodhisattva ideal of benefiting all sentient beings. With these new translation terms, he is apparently trying to reach a wider public. As an unintended consequence, the 2004 translation is harder to use by students of Buddhism who are accustomed to more standard translations of Buddhist terms, and who may well make up the book’s largest readership. Ironically, 27 years earlier in 1977, Thurman had rather harshly reviewed the translation of Longchenpa’s text, Kindly Bent to Ease Us, made by Herbert Guenther, who has become well-known for his unique choices of translation terms: “Unfortunately, Guenther ruins the whole thing, shrouding the jewel of the original with his own intellectual obscurities so that we catch only an occasional glint of its brilliance.”10 It is certainly true that a glossary would have helped the 2004 translation immensely, and one will no doubt be added in a future edition. As the first volume in the Tanjur Translation Initiative, this book was published under more difficult circumstances than normal, and subsequent volumes in this series do have glossaries.
We may now turn to a few example verses from the two translations. Kamaleswar Bhattacharya picked out verse 9.23 (or 10.23) as a key verse with which to open his book, L’Ātman-Brahman dans le Bouddhisme ancien (The Ātman-Brahman in Ancient Buddhism). This verse pertains to the question of the ātman or “self,” whose denial is considered to be one of the defining characteristics of Buddhism. The Sanskrit is given below from Lévi (1907) as corrected by Nagao (1958) and Funahashi (1985), and also in a footnote in the 2004 translation. As noted by Bhattacharya (2001, p. 6), it turns out that the incorrect Sanskrit reading found in Lévi’s 1907 edition, nairātmyānmārgalābhataḥ, is a silent emendation by Lévi himself. His manuscript had it correct except for a missing “r” (a small stroke under the “ga”), which threw him off the right track. Bhattacharya reproduces the actual manuscript folio that Lévi used, showing the reading, nairātmyātmāgalābhataḥ (at the very beginning of that folio). The Tibetan is given below from the Comparative Tengyur published in China (vol. 70, 2001, p. 823, lines 4-5, text of the verses only, having the present form ’gyur for the last syllable, and pp. 1196-1197, text of the verses with commentary, having the past form gyur for the last syllable, which I adopt in agreement with the Sanskrit past form gata).
śūnyatāyāṃ viśuddhāyāṃ nairātmyâtmâgra-lābhataḥ |
buddhāḥ śuddhâtma-lābhitvāt gatā ātma-mahâtmatām || 9.23 ||
stong pa nyid ni rnam dag na || bdag med mchog gi bdag thob pas ||
sangs rgyas dag pa’i bdag thob phyir || bdag nyid chen po’i bdag tu gyur ||
9.23. In pure voidness buddhas achieve the supreme self of selflessness, and realize the spiritual greatness of the self by discovering the pure self. (2004 translation)
10.23. Within pure emptiness,
The buddhas achieve the supreme self of selflessness.
Thus they achieve the pure self,
And are hence the self of great beings. (2014 translation)
First, we see that both translations use “selflessness” for nairātmya (Tib. bdag med). The word selflessness in English has always meant unselfishness or altruism. Here it has been employed to mean something very different, the Buddhist teaching of the “absence of a self” in persons (pudgala-s), and according to Mahāyāna Buddhism, also in things or phenomena (dharma-s). If you are “in the loop,” if you are among those who have read a number of modern books on Buddhism, you will know this meaning and usage of the word selflessness. If you are not in the loop, this translation of nairātmya will make little sense.
Another translation of this verse, one that follows the Sanskrit very closely, was made by Paul Griffiths in a 1990 article (p. 52):11
“In pure emptiness,
By obtaining the supreme self which is without self,
Buddhas arrive at the great-selfed self
As a result of obtaining the pure self.”
The 2014 translation says that the buddhas “are hence the self of great beings.” While the Tibetan translation allows this English translation, the Sanskrit, both of the verse and of the commentary, does not. In the Tibetan words bdag nyid chen po, taken as “great being,” the nyid actually translates the Sanskrit abstract suffix -tā, “-ness,” on mahātmatā, literally “great-self-ness,” or “great-selfed” in the Griffiths translation, or just “greatness” in the 2004 translation.
Vasubandhu’s commentary tells us that this verse is about the highest self (paramâtman) of the buddhas in the uncontaminated (anāsrava) realm (dhātu, here Tib. dbyings, and also may be translated as space or element). Vasubandhu also tells us that it is the self (ātman) of the buddhas in the sense of “inherent nature” (svabhāva), important because both ātman and svabhāva are otherwise denied in Mahāyāna Buddhism. Unless we know that “intrinsic reality” translates svabhāva in the 2004 translation, as a glossary would tell us, we would miss this. Here is Vasubandhu’s commentary on this verse as found in the 2004 translation, preceded by the Sanskrit (from Funahashi’s edition, with one missing diacritic restored by me, otherwise agreeing with Lévi’s edition) and Tibetan (Comparative Tengyur, vol. 70, p. 1197, showing five variant readings, of which I cite only one):
tatra cânāsrave dhātau buddhānāṃ paramâtmā nirdiśyate | kiṃ kāraṇaṃ | agra-nairātmyâtmakatvāt | agraṃ nairātmyaṃ viśuddhā tathatā sā ca buddhānām ātmā svabhāvârthena tasyāṃ viśuddhāyām agraṃ nairātmyam ātmānaṃ buddhā labhante śuddhaṃ | ataḥ śuddhâtma-lābhitvāt buddhā ātma-māhātmyaṃ prāptā ity anenâbhisaṃdhinā buddhānām anāsrave dhātau paramâtmā vyavasthāpyate |
zag pa med pa’i dbyings de la sangs rgyas rnams kyi bdag nyid kyi mchog ston te | ci’i phyir zhe na | bdag med pa mchog gi bdag nyid kyi phyir ro || bdag med pa mchog ni de bzhin nyid rnam par dag pa’o || de yang ngo bo nyid kyi don gyis sangs rgyas rnams kyi bdag yin no || de rnam par dag na sangs rgyas rnams kyis bdag med pa mchog gi bdag nyid dag pa ’thob po || de bas na sangs rgyas rnams kyi dag pa’i bdag thob pa’i phyir bdag nyid chen po’i bdag tu gyur pa yin te | dgongs pa ’di* ni zag pa med pa’i dbyings la sangs rgyas rnams kyi bdag gi mchog rnam par ’jog go ||
*’dis in the Peking and Narthang editions.
“This shows the supreme self of the buddhas in the uncontaminated realm. Why? Because hers is the self of supreme selflessness. Supreme selflessness is completely pure suchness, and that is a buddha’s ‘self,’ in the sense of ‘intrinsic reality.’ When this is completely pure, buddhas attain superior selflessness, a pure self. Therefore, by attaining a pure self buddhas realize the spiritual greatness of self. Thus it is with this intention that buddhas are declared to have a supreme self in the uncontaminated realm.”
Khenpo Shenga’s commentary is here quite brief, extracting only a couple of points from Vasubandhu’s commentary. As found in the 2014 translation, Khenpo Shenga’s commentary on this verse follows. Words quoted from the verse itself are put in bold, a helpful feature.
“Within pure emptiness, the buddhas achieve the suchness that is the supreme self of selflessness. Thus they achieve the supremely pure self, and hence they are the self that is the realization of great beings.”
Ju Mipham’s commentary is also comparatively brief here, making up less than half a page in the 2014 translation. Sthiramati’s commentary on this verse makes up two full pages in the English translation of chapter 9 of this commentary that is included in Cuong Tu Nguyen’s 1990 Harvard PhD. thesis (attached, see link in footnote).12 As a comparison of these commentaries on this verse will show, Mipham here takes little from Sthiramati, but instead comments more in accordance with the “Great Madhyamaka” ideas that form the basis of the Ri-mé or “non-sectarian” movement. Mipham was one of the major teachers of this late nineteenth-century movement in Tibet. Here is Mipham’s commentary on this verse as found in the 2014 translation:
“The pure and natural luminosity of emptiness is completely free from the self-manifestation of the adventitious defilements. In the absence of the twofold self of persons and phenomena, this is the actual nature of things, the supreme nature of the abiding reality, the intrinsic nature or essence itself. In achieving this, the buddhas have achieved a nature that is of complete purity. Thus, [to actualize] the suchness that is the unmistaken way things are is to be ‘the self of great beings.’ This self is not the same as the conceived object that is involved when apprehending the twofold self because such a self has no bearing on things as they are. The buddhas, however, have actualized the unmistaken abiding reality, which is the suchness of the twofold selflessness, free from the extremes of existence and nonexistence. That is the supreme self—‘the self of great beings.’”
The next example is from the third chapter (or fourth in the Tibetan translation). This is the first chapter on a Buddhist doctrinal topic, after the introductory chapter(s) and the chapter on going for refuge. Its topic is the gotra (Tib. rigs), a term that is very hard to translate adequately into English. David Seyfort Ruegg has distinguished three main meanings in Buddhist usage: 1. mine, matrix; 2. family, clan, lineage; 3. germ, seed.13 It is translated as “spiritual gene” in the 2004 translation, although in a 1979 draft of this translation that I have access to, it was translated as “heritage.” It is translated as “potential” in the 2014 translation. (It is left untranslated in the 1992 translation.) Verse 4 of this chapter gives its defining characteristics. The Sanskrit is given from Funahashi’s edition, agreeing with Lévi’s edition. The Tibetan is given from the Comparative Tengyur, vol. 70, p. 810, lines 6-8, where the text of only the verses has a variant reading, and from vol. 70, p. 1152, lines 9-11, where the text of the verses with commentary has another variant reading. I have ignored a third variant reading that is obviously an error.
prakṛtyā paripuṣṭaṃ ca āśrayaś câśritaṃ ca tat |
sad asac câiva vijñeyaṃ guṇôttāraṇatârthataḥ || 3.4 ||
rang bzhin dang ni rgyas pa dang || de ni rten dang brten pa dang ||
yod med nyid* dang yon tan ni** || sgrol ba’i don du shes par bya ||
*gnyis in the Der-ge and Co-ne editions of the text of the verses with commentary.
**dang in the Peking and Narthang editions of the text of the verses only.
3.4 “Natural, developed, support, supported, existent and nonexistent; it is to be understood in the sense of “delivering excellences.” (2004 translation)
4.4 “The natural and the developed
Are the support and the supported.
Present while not present,
It should be known to mean “freeing qualities.” (2014 translation)
Vasubandhu’s commentary explains this verse, as found in the 2004 translation, preceded by the Sanskrit (from Funahashi’s edition, with one missing diacritic restored by me, otherwise agreeing with Lévi’s edition) and Tibetan (Comparative Tengyur, vol. 70, p. 1152, ignoring two variant readings that are obvious scribal errors):
etena catur-vidhaṃ gotraṃ darśayati | prakṛti-sthaṃ samudānītam āśraya-svabhāvam āśrita-svabhāvaṃ ca tad eva yathā-kramaṃ | tat punar hetu-bhāvena sat phala-bhāvenâsat | guṇôttāraṇârthena gotraṃ veditavyaṃ guṇā uttaranty asmād uddhavantîti kṛtvā |
’dis ni rigs rnam pa bzhi ston te | rang bzhin du gnas pa dang | yang dag par bsgrubs pa dang | rten gyi ngo bo nyid dang | brten pa’i ngo bo nyid de de dag nyid dang go rims bzhin no || de ni rgyu’i dngos por yod do || ’bras bu’i dngos por med do || rigs ni yon tan sgrol ba’i don du yang rig par bya ste | ’di las yon tan sgrol zhing ’byung ba’i phyir ro ||
“This shows the spiritual gene to be fourfold: existing by nature, being developed, having the nature of a support, and having the nature of the supported, respectively. It exists as a cause, it does not exist as an effect. The spiritual gene is to be understood in the sense of ‘delivering excellences’; because excellences are delivered—that is, emerge—from it.”
The Mahāyāna-sūtrālaṃkāra chapter and Vasubandhu’s commentary thereon, consisting of thirteen verses, give the gotra teachings briefly. They are given more extensively, and in prose, in the Bodhisattva-bhūmi, where they form the first chapter. Thurman writes in his Introduction (p. xxxv): “The BBh [Bodhisattva-bhūmi] follows the pattern of the MSA [Mahāyāna-sūtrālaṃkāra] very closely, which is why I consider it to be Asaṅga’s own ‘meaning-’ or ‘depth-commentary’ (Tib. don ’grel) on the text.” It certainly does give the teachings in more depth. This is especially true of the tattvārtha or “reality” chapter. This is the sixth chapter (or seventh in the Tibetan translation) of the Mahāyāna-sūtrālaṃkāra, consisting of only ten verses. It is the fourth chapter of the Bodhisattva-bhūmi, consisting of twenty-one pages in the 1937 Unrai Wogihara edition (pp. 37-57), and of fifteen pages in the 1966 Nalinaksha Dutt edition (pp. 25-39) (both posted on this website under “Sanskrit Texts”). The central theme of this chapter in the Bodhisattva-bhūmi is the vastu, the “thing” in itself. The vastu is not even mentioned in this chapter of the Mahāyāna-sūtrālaṃkāra. H. P. Blavatsky in a private letter of 1886, describing The Secret Doctrine that she was then writing, linked the “Book of Dzyan” with the secret book of Maitreya Buddha. By contrast, she referred to the known five books of Maitreya, which are written in verse, as blinds:
“I have finished an enormous Introductory Chapter, or Preamble, Prologue, call it what you will; just to show the reader that the text as it goes, every Section beginning with a page of translation from the Book of Dzyan and the Secret Book of ‘Maytreya Buddha’ Champai chhos Nga (in prose, not the five books in verse known, which are a blind) are no fiction.”14
Blavatsky’s description of the known verse works of Maitreya, including the Mahāyāna-sūtrālaṃkāra, as a “blind” seems to be fitting when we compare it to the much more detailed teachings in the prose Bodhisattva-bhūmi. Nonetheless, even a “blind” (if it is such), contains important teachings, however brief. The Mahāyāna-sūtrālaṃkāra in this tattvārtha chapter speaks of the dharma-dhātu beyond mind in verses 7-8. These are key verses for the Yogācāra school of Buddhism, often held to teach “mind-only” (citta-mātra). Here are these verses in the two translations, preceded by the Sanskrit (from Lévi’s edition, transliterated and hyphenated by me) and Tibetan (Comparative Tengyur, vol. 70, p. 816, lines 6-11, ignoring one variant reading, and p. 1174, lines 4-9, also ignoring one variant reading, a different one). Note that dharma-dhātu is translated in the 2004 translation as the “ultimate realm,” and in the 2014 translation as the “basic field of phenomena.”
arthān sa vijñāya ca jalpa-mātrān saṃtiṣṭhate tan-nibha-citta-mātre |
pratyakṣatām eti ca dharma-dhātus tasmād viyukto dvaya-lakṣaṇena || 6.7 ||
nâstîti cittāt param etya buddhyā cittasya nâstitvam upaiti tasmāt |
dvayasya nâstitvam upetya dhīmān saṃtiṣṭhate ’tad-gati-dharma-dhātau || 6.8 ||
de yis brjod pa tsam du don rig nas || der snang sems tsam la ni yang dag gnas ||
de nas chos dbyings gnyis kyi mtshan nyid dang || bral ba mngon sum nyid du rtogs par ’gyur ||
sems las gzhan med par ni blos rig nas || de nas sems kyang med pa nyid du rtogs ||
blo dang ldan pas gnyis po med rig nas || de mi ldan pa’i chos kyi dbyings la gnas ||
6.7. And once aware that objects are mere verbalizations she securely dwells in the realm of mind alone with such (objective) appearance. Then she realizes intuitively that the ultimate realm is (immanently) present, free of the nature of duality.
6.8 Realizing intellectually that there is nothing apart from mind, she understands then that mind (itself) has no (ultimate) existence. Understanding that duality has no existence, such a genius dwells in the ultimate realm which has no (duality). (2004 translation)
7.7 Hence, knowing objects to be mere expressions,
The bodhisattva recognizes that such appearances are mind only,
And then realizes the basic field of phenomena,
Free from the characteristics of duality, in direct perception.
7.8 Becoming aware that there is nothing apart from the mind,
The bodhisattva also realizes that the mind does not exist at all.
Having seen that the two do not exist, the intelligent one abides
In the basic field of phenomena, which does not contain them. (2014 translation)
As we see, following upon the idea that nothing exists other than mind (nâstîti cittāt param), these verses say the bodhisattva realizes that the mind does not exist (cittasya nâstitvam). The Mahāyāna-sūtrālaṃkāra is a fundamental text of the Yogācāra school, a school that is widely held to teach the existence of “mind-only” (citta-mātra), and thus is also called the Cittamātra school. The “Great Madhyamaka” tradition claims the five treatises of Maitreya as its source texts, saying that these texts do not teach “mind-only”; but rather they teach that mind, like all other phenomena, does not ultimately exist. These verses from the Mahāyāna-sūtrālaṃkāra would be an important source reference in support of this assertion. Interestingly, although Mipham as a Ri-mé teacher is a major exponent of the Great Madhyamaka tradition, he does not bring out this point in his commentary on these verses.
The example verses quoted so far were chosen to illustrate important ideas found in the Mahāyāna-sūtrālaṃkāra. They have not much illustrated the differences between the two translations in translation terminology. For this, we may look at verse 9.9 (or 10.9). Before doing so, we must note that the Tibetan translation of this verse differs from Lévi’s Sanskrit edition in two places. Neither Lévi in his corrections published long posthumously in 2001 nor Nagao in his corrigenda published in 1958 suggested emendations to this Sanskrit verse. So we are glad to see that the 2004 translation in a footnote (p. 76, fn. 16) gives emendations to this Sanskrit verse in three places, even though none of these three emendations fit the meter, and the third of these is unnecessary.15 The second of these emendations concerns the word ’jig tshogs found in the Tibetan translation. This is the standard translation of the Sanskrit word satkāya, which is not found in Lévi’s Sanskrit edition. However, Funahashi in his 1985 revised edition of this chapter shows that five Nepalese manuscripts do have satkāya here. The words sarvarakṣāpayānaṃ in Lévi’s edition thus should be sarvasatkāyayāna as in Funahashi’s edition. This also fits the meter. I give his revised text of this verse below. The remaining emendation to this Sanskrit verse is not so easy to ascertain.16 I give the Tibetan from the text of only the verses in the Comparative Tengyur (vol. 70, p. 821, lines 13-16), which has one significant variant reading, and two insignificant ones that I have ignored. Likewise the text of the verses with commentary has two variant readings that I have ignored (vol. 70, p. 1192, lines 18-21).
śaraṇam anupamaṃ tac chreṣṭha-buddhatvam iṣṭaṃ janana-maraṇa-sarva-kleśa-pāpeṣu rakṣā |
vividha-bhaya-gatānāṃ sarva-satkāya-yāna-pratata-vividha-duḥkhâpāya-nôpāya-gānāṃ || 9.9 ||
sangs rgyas nyid de skyabs ni dpe med mchog tu ’dod ||
sna tshogs ’jigs gyur ’jig tshogs kun dang theg pa dang ||
ngan song rnam mang sdug bsngal thabs min song ba rnams ||
skye dang ’chi dang nyon mongs ngan song* kun las srung ||
*las rnams in the Peking and Narthang editions of the text of the verses only.
9.9 Supreme buddhahood is accepted as the incomparable refuge. It grants protection amidst births and deaths, amidst all addictions and hellish migrations, for all those who have fallen into various dangers, materiality, (inferior) vehicles, unremitting suffering of various kinds, hellish rebirths, and unliberating arts. (2004 translation)
10.9 The refuge of buddhahood is held to be incomparably supreme,
For it protects against the different fears, all of the transitory collection, the vehicles,
The numerous sufferings of the lower realms, the pursuit of nonmethods,
Birth, death, afflictions, and the lower realms. (2014 translation)
We notice in the 2004 translation three unique translation terms:
(1) “addictions” for kleśa-s (Tib. nyon mongs), translated as “afflictions” in the 2014 translation. As said above, “afflictions” (or “mental afflictions”) has now become a frequent translation for kleśa-s in Buddhist texts, as has “afflictive emotions.” I have also seen “mental and moral afflictions.” These translations are based on the etymological and literal meaning of kleśa as “affliction.” Interestingly, “affliction” was also the earliest English translation of kleśa, found in James R. Ballantyne’s 1852 and 1853 translation of Yoga-sūtra books 1 and 2 (where they are enumerated at 2.3), and adopted by many other translators of this Hindu text up to the present. In previous translations of Buddhist texts this term was often given more descriptive translations such as “defilements,” “moral defilements,” “defiled emotions,” “passions,” etc. The kleśa-s are desire, hatred, delusion, pride, ignorance, wrong views, doubt, etc.
(2) “materiality” for sat-kāya (Tib. ’jig tshogs), translated as “the transitory collection” in the 2014 translation. The term “transitory collection” is not uncommon in English translations made from the Tibetan, since it is a translation of the Tibetan translation, ’jig tshogs, which in turn is a literal translation of the Sanskrit term, sat-kāya, as it is explained in Buddhist texts.17 This term is associated with the basic Buddhist teaching of ātma-dṛṣṭi, the “(false) view of self.” The transitory or perishable collection or aggregation refers to the body, feelings, thoughts, etc. (the skandha-s), that together make up a person, and which is falsely regarded as a permanent self. This term is therefore often given more descriptive translations. Thus, it is translated as “false views of self” in Cuong Nguyen’s translation of Sthiramati’s commentary on this verse (p. 359). Incidentally, Sthiramati takes the word “all” (sarva) with it in this verse, “all false views of self.”
(3) “unliberating arts” for na upāya (here used in a compound for anupāya in order to fit the meter, Tib. thabs min), translated as “nonmethods” in the 2014 translation. From the term “nonmethods” we can easily derive “methods,” a common translation of upāya, which is also often translated as “means.” Likewise, from “unliberating arts” we can derive “liberating arts,” which is used throughout the 2004 translation for upāya. This term is frequently seen with prajñā in the contrasting and complementing pair, “wisdom and means.” It is also frequently seen with kauśalya in the phrase, “skill in means.”
Another unique translation term found in the 2004 translation is “theology” for tarka (Tib. rtog ge), translated as “logic” in the 2014 translation and elsewhere. Thus, we read in verse 1.12 (Lévi Sanskrit edition) or 1.6 (2004 translation) or 2.6 (2014 translation):
1.6. Theology is dependent, indefinite, non-comprehensive, superficial, tiresome, and the resort of the naïve. Thus, this (universal vehicle) is not within its scope. (2004 translation)
2.6. Logic is dependent, uncertain,
Incomprehensive, relative, and tiresome.
It is held to be reliable by the childish,
And this is, therefore, not within the domain. (2014 translation)
As noted above, the anomalous use of “selflessness” has become accepted Buddhist jargon for those in the know. In combination, this leads to another unique translation term found in the 2004 translation, one that may take more than being in the loop to understand. In Vasubandhu’s commentary on verse 4.14 we read, “there is equanimity towards all things due to the understanding of objective selflessness.” In normal English, “objective selflessness” would mean “unbiased altruism” or “impartial unselfishness,” and this is something we might expect from a bodhisattva who has equanimity towards all things. Now that we are in the loop, however, we know that “selflessness” here means “absence of self,” not “altruism” or “unselfishness.” So we next need to determine what “objective absence of self” might mean. To do this, we must have studied Mahāyāna Buddhism long enough to know that it teaches two kinds of “absence of self”: that of persons and that of phenomena or things. We can then see that “objective selflessness” must mean “absence of self in objects,” i.e., in things or phenomena. Without such a background, I do not think that this phrase would be understood to mean this. This phrase is found in the 2014 translation as “selflessness of phenomena.” The word “phenomena,” too, has become accepted Buddhist jargon. A Christian theologian pursuing interfaith studies may not find either of these translations to be very comprehensible.
While the 2014 translation normally uses translation terminology that has now come in to common use, it does use a few uncommon or unique translation terms. These are, perhaps, harder to recognize in this translation because they are unexpected there. For example, it uses “intrinsic nature” for dharmatā (Tib. chos nyid), a translation term that elsewhere almost always translates svabhāva (Tib. ngo bo nyid, rang bzhin). Thus, in verse 2.5 (= 1.11 in the Lévi Sanskrit edition), we read: “It [the Great Vehicle] does not conflict with the intrinsic nature”; while in the 2004 translation (= 1.5) we find, “it [the universal vehicle] does not run counter to actual reality.” The term “actual reality,” like “true reality,” is within the norm for dharmatā, whose most common translation is “true nature.”
Also unexpected in the 2014 translation is the translation of saṃjñā (Tib. ’du shes) as “identification.” There we read in verse 10.47: “When the identification of space has transformed, whatever is wished for manifests.” In the 2004 translation we find a more common translation of saṃjñā as “conception” in verse 9.47: “In the transmutation of the conception of space, highest mastery is attained.”
Likewise the translation of vijñapti (Tib. rnam par rig pa) as “awareness” in the 2014 translation is unexpected and therefore apt to be confusing. There we read in verse 12.24: “The causes of delusion and delusion are held to be awareness of form and awareness without form.” In the 2004 translation we find a more common translation of vijñapti as “idea” in verse 11.24: “The cause of error and error itself are considered to be the idea of matter and the idea of nonmateriality (respectively).”
These few unusual translation terms in the 2014 translation are very much the exception, and I call attention to them for the reason that they are unexpected there. The vast majority of the translation terms in the 2014 translation are ones that would be expected. Moreover, it does have a glossary, even though such glossaries are necessarily selective. Thus, for example, it leaves out “transcendence of suffering,” for nirvāṇa. Also, the English-Tibetan Glossary, through some glitch, omits all the words starting with “s”.
In one case, both translations use uncommon or unique translation terms. For jñāna (Tib. ye shes), common translations are “knowledge,” “wisdom,” “gnosis,” etc. The 2004 translation uses “intuition” for it, and the 2014 translation uses “wakefulness” for it. While such translations can provide helpful insights into the meaning of the original term, they can also make it harder to get the intended meaning, as may be seen in the following verse:
9.34. Just as clouds and so forth are thought to obscure the rays of sunlight, so the deficiencies of beings obscure the buddhas’ intuitions. (2004 translation)
10.34. It is held that the rays of the sun
Are obscured by things such as clouds.
In the same way, the wakefulness of the buddhas
Is obscured by the flaws of sentient beings. (2014 translation)
Both of these translations give the impression that the deficiencies or flaws of sentient beings interfere with the insights or awareness that the buddhas would otherwise have. Of course, the intended meaning is that the deficiencies or flaws of sentient beings interfere with their own realization of the wisdom or knowledge or gnosis possessed by the buddhas. This is clear in the translation of this verse by Cuong Nguyen:
9.34. Just as clouds and the like obstruct the sunlight, so the faults of sentient beings block the Buddhas’ wisdoms. (1990 thesis, p. 393)
The renowned accuracy of the Tibetan translations in very closely following the Sanskrit originals goes hand in hand with their use of standardized translation terminology. This was implemented quite early by royal decree, and was used throughout the entire body of Buddhist texts. This standardized translation terminology allowed Tibetans to know that chos is always dharma, for example, no matter in what text or who translated it. We do not have this in our English translations today, nor are we likely to, because of our individualistic natures. Thurman has noted that the śāstra texts comprising the Tengyur, of which the Mahāyāna-sūtrālaṃkāra is one, are scientific treatises (pp. ii, vii, xvii). While their primary field is not the physical realm, as is that of the modern sciences of biology, chemistry, physics, etc., what they expound are similarly sciences that require the use of precise technical terms. Lacking standardized translation terms that all can agree on, we are obliged to add glossaries, or to add the Sanskrit terms in parentheses (as done by Étienne Lamotte in his valuable translations), or even to add the whole Sanskrit text (as is now frequent in translations of Hindu texts published in India). Things were different when the Buddhist texts were translated from Sanskrit into Tibetan. The use of standardized translation terminology, along with the literal accuracy of the Tibetan translations, together resulted in the most precise transferal of a body of religious knowledge from one language to another known to history.
In conclusion, the two translations of the Mahāyāna-sūtrālaṃkāra complement each other in important ways. The use of more standard translation terminology makes the 2014 translation more understandable, while the use of the Sanskrit original makes the 2004 translation more accurate. No serious student can afford to be without either of them.
1. The Works of Maitreya: English Translations, p. 7. Eastern Tradition Research Institute Bibliographic Guides, 2007: http://easterntradition.org/etri%20bib-maitreya.pdf.
2. See Sylvain Lévi’s Avant-propos to his 1907 Sanskrit edition. This edition is posted here under “Sanskrit Texts,” then “Sanskrit Buddhist Texts,” then Mahāyānasūtrālaṃkāra: mahayana_sutralamkara_1907.pdf. An English translation of this Avant-propos was made by Umesh Jha and published, along with the French, as “A Rendition of Lévi’s Preface to the Sūtrālaṃkāra,” Bulletin of the Mithila Institute of Post-Graduate Studies and Research in Sanskrit Learning, Darbhanga, Vols. IV-VI, Sept. 1968-Sept. 1970, pp. 202-209, here attached as: Mahayana-sutralamkara, Levi’s Preface, Eng. The relevant portion is also quoted in French and translated into English by Kamaleswar Bhattacharya in his 2001 article, pp. 5-6 and fn. 5; for the full title and link, see note 8 below.
3. As Kazuo Kano informs us in his 2012 article, “Eight Folios from a Sanskrit Manuscript of the Mahāyānasūtrālaṃkārabhāṣya from Ngor Monastery: Diplomatic and Critical Editions on X.9-XI.3,” p. 33. See note 9 below for link.
4. Gadjin M. Nagao, “Corrigenda of the Text Edited by Professor Sylvain Lévi,” in Index to the Mahāyāna-sūtrālaṃkāra (Sylvain Lévi Edition), Part One: Sanskrit-Tibetan-Chinese, pp. xi-xxii (Tokyo, 1958), here attached as: Mahayana-sutralamkara corrigenda Nagao 1958.
5. The two additional Sanskrit manuscripts that were brought to Japan and are kept in the Ryūkoku University Library were first reported on and studied by Shōko Takeuchi in his Japanese language article, “On Mahāyānasūtrālaṃkāra—brought by Ōtani Mission,” Ryūkoku Daigaku Ronshū, no. 352, Aug. 1956, pp. 72-87, here attached as: Mahayanasutralamkara brought by Otani Mission, Takeuchi 1956. Besides being consulted by Nagao, these two manuscripts were also used by Takanori Umino, in his English language article, “Corrections of the Mahāyānasūtrālaṃkāra XI. 35,” Journal of Indian and Buddhist Studies, vol. 22, no. 1, Dec. 1973, pp. 513-508 (20-25), here attached as: Mahayanasutralamkara Corrections of XI.35, Takanori 1973.
6. Two of these additional Sanskrit manuscripts from the Nepal National Archives were compared with Lévi’s edition by Risho Hotori, who published a “Concordance of the Sanskrit Edition and Two Manuscripts of the Mahāyānasūtrālaṃkāra,” in Tetsugaku Nempō, no. 43, Feb. 1984, pp. 83-90, here attached as: Mahayanasutralamkara Concordance Two Manuscripts, Hotori 1984. These two manuscripts were used by Gadjin Nagao, along with the two from the Ryūkoku University Library, for his English translation of chapter 17, verses 29-64, with revised Sanskrit edition and list of corrections to Lévi’s edition, published as “The Bodhisattva’s Compassion Described in the Mahāyāna-sūtrālaṃkāra,” in Wisdom, Compassion, and the Search for Understanding (University of Hawai’i Press, 2000), pp. 1-38, here attached as: Mahayana-sutralamkara 17.29-64 Eng. Skt. Nagao 2000.
7. Naoya Funahashi, Mahāyānasūtrālaṃkāra (Chapter I, II, III, IX, X), Revised on the basis of Nepalese manuscripts (Tokyo, 1985). This is posted here under “Sanskrit Texts,” then “Sanskrit Buddhist Texts,” then Mahāyānasūtrālaṃkāra: mahayana_sutralamkara_partial_1985.pdf.
8. Kamaleswar Bhattacharya, “For a New Edition of the Mahāyānasūtrālaṃkāra, Journal of the Nepal Research Centre, vol. XII, 2001, pp. 5-16, here attached as: Mahayanasutralamkara, For a New Edition of, Bhattacharya 2001.
9. Kazuo Kano has kindly posted his many valuable articles at Academia.edu (https://koyasan-u.academia.edu/KazuoKano). This is very helpful because Japanese academic publications are not easily accessible here in the U.S.A., for example. Besides his article listed in note 3 above, his three other articles on the Mahāyāna-sūtrālaṃkāra are: “Palm-leaf Manuscript of the Mahāyānasūtrālaṃkāra from Ngor Monastery—Folio 27: XI.14-27—,” “The Sanskrit Manuscript of the Mahāyānasūtrālaṃkārabhāṣya from Ngor Monastery: Diplomatic Edition on XVII.37-39,” and “Vairocanarakṣita’s Glosses of the Mahāyānasūtrālaṃkārabhāṣya Chapter 17.” In his article listed in note 3 above (pp. 36-37) he gives information about other Mahāyāna-sūtrālaṃkāra manuscripts in Tibet. As access to these becomes possible, we may hope to eventually have a very accurate Sanskrit edition of this text. From access to an incomplete related text, the Sūtrālaṃkāra-paricaya, Ye Shaoyong was able to recover three verses, 2.9-11, that are absent in Lévi’s edition due to a missing folio: “Three Verses of the Mahāyānasūtrālaṃkāra Missing in Sylvain Lévi’s Edition,” Journal of Sino-Western Communications, vol. 5, no. 1, July 2013, pp. 218-224, here attached as: Mahayanasutralamkara, three missing verses.
10. Robert A. F. Thurman, review of Herbert V. Guenther, Kindly Bent to Ease Us, in Harvard Journal of Asiatic Studies, vol. 37, no. 1, June 1977, pp. 222-228, here attached as: Thurman review of Guenther Kindly Bent to Ease Us.
11. Paul J. Griffiths, “Painting Space with Colors: Tathāgatagarbha in the Mahāyānasūtrâlaṅkāra-Corpus IX.22-37,” in Buddha Nature: A Festschrift in Honor of Minoru Kiyota, pp. 41-63 (Tokyo, 1990), here attached as: Mahayanasutralamkara 9.22-37, Tathagatagarbha in, Griffiths 1990.
12. Cuong Tu Nguyen, Sthiramati’s Interpretation of Buddhology and Soteriology, Harvard University PhD. thesis, 1990, pp. 379-383, including verse 9.23, here attached as: Mahayana-sutralamkara Sthiramati comm. 9.23 Nguyen trans.
13. D. Seyfort Ruegg, “The Meanings of the Term Gotra and the Textual History of the Ratnagotravibhāga,” Bulletin of the School of Oriental and African Studies, vol. 39, 1976, p. 354. This article is posted here under “References,” then “Studies,” then “Dhatu — Gotra (Eleven articles),” of which it is the fifth article, pp. 28-40 of that PDF.
14. The Letters of H. P. Blavatsky to A. P. Sinnett, London, 1925, p. 195.
15. The third emendation to this Sanskrit verse given in a footnote in the 2004 translation (p. 76, fn. 16) is pratatavividhaduḥkhāpāyānupāgānāṃ for Lévi’s pratatavividhaduḥkhāpāyanopāyagānāṃ. Apparently this emendation is itself a typographical error, since it lacks a syllable and eliminates the word upāya, for which we have its standard translation thabs in the Tibetan text. Probably the intended emendation was pratatavividhaduḥkhāpāyānupāyagānāṃ. In any case, it is unnecessary. The use of the Sanskrit word “na” in a compound in order to fit the meter, here nopāya instead of anupāya, is not uncommon.
16. The remaining emendation to this Sanskrit verse concerns the word pāpa (in kleśapāpeṣu), for which the Tibetan translation (in the Der-ge edition used in the 2004 translation, signified by “D” but not in the list of abbreviations) has ngan song, the standard translation of the Sanskrit word apāya. Since the letter “p” looks almost like the letter “y” in Sanskrit manuscripts, this allows the apparently easy emendation kleśāpāyeṣu, as given in the 2004 translation footnote. However, the long “ā” resulting from merging kleśa and apāya goes against the meter. The printed reading, kleśapāpeṣu, fits the meter, and is apparently found in all of the several Nepalese manuscripts collated by Funahashi. Since apāya is mentioned later in this verse, there would be no need to also have it here. Then, there is a variant reading in the Tibetan translation of this verse in the text giving the verses alone (but not in the text giving the verses and commentary together, Comparative Tengyur, vol. 70, p. 1192, lines 18-21). For ngan song in the Der-ge and Co-ne editions, the Peking and Narthang editions have las rnams (Comparative Tengyur, vol. 70, p. 821, line 16). The Tibetan word las translates the Sanskrit word karma (the rnams is the plural marker). This indicates that the Sanskrit manuscript(s) used for the Peking/Narthang edition had kleśakarmeṣu here. This also fits the meter. The Tibetan translation of Sthiramati’s commentary here has ngan song, seeming to confirm apāya, but it explains las, karma, in conjunction with nyon mongs, kleśa, the “mental/moral afflictions.” So we do not know whether Maitreya here spoke of protection from pāpa, “sins,” apāya, “bad rebirths,” or karma, “actions.”
17. The Sanskrit term sat-kāya looks like it should mean “real body,” or “truly existing body.” However, as explained in Buddhist texts such as the Abhidharma-kośa-bhāṣya (5.7), here sat means sīdati. That is, it comes from the root sad, meaning “to break, decay, perish.” It is not the present participle or noun sat from the root as, meaning “existing, truly existing, real.” Also, here kāya is taken in its meaning, “assemblage, aggregation, collection” rather than “body.” The Tibetan ’jig tshogs is a literal translation of this, meaning “disintegrating collection,” and thus is taken as “transitory collection.”
Additional note: A four-language electronic edition of the Mahāyāna-sūtrālaṃkāra is available at the University of Oslo Bibliotheca Polyglotta website. It includes Sanskrit, Chinese, Tibetan, and French. It is very convenient, but must be used with caution at present. This is because, judging by the many typographical errors, it does not seem to have been proofread. It can be found at: http://www2.hf.uio.no/polyglotta/index.php?page=fulltext&view=fulltext&vid=85&cid=182062&mid=283928&level=1